A COMPARISON OF FORECASTING METHODS FOR NUMBER OF TOURISTS IN PRASAT MUANGTAM, PRAKHONCHAI DISTRICT, BURIRAM PROVINCE.

Abstract

The purposes of this research were to compare suitable forecasting methods, and to construct the Mathematical Model for number of tourists in Prasathin Muang Tam, Prakhon Chai District, Buriram Province for 132 values from October 2007 to September 2018.  The data were separated into two sets: the first set consisted of 120 values from October 2007 to September 2017 which was used for study five methods of forecasting models, namely, Weighted Moving Average, Brown’s One-Parameter Method, Holt’s Two-Parameter Method, Winter’s Three-Parameter Trend & Seasonality Method, and Decomposition Method. Another set, the last 12 values from October 2017 to September 2018 was used for the comparison of predictive values ​​obtained, and the selection of accurate models. The criteria used was the average absolute error average, percentage, and absolute error. Research findings indicate that for all the forecasting methods, Winter’s Three-Parameter Trend & Seasonality Method is the most suitable for this prediction in the number of tourists in Prasat Muang Tam. The mathematical model for foreclosing was successfully developed.